- Email: [email protected]
Integration of operational research into National Tuberculosis Control Programmes
Tuberculosis (2003) 83, 143–147
Tuberculosis www.elsevierhealth.com/journals/tube
Integration of operational research into National Tuberculosis Control Programmes Anthony D. Harries* Malawi National Tuberculosis Control Programme, Community Health Science Unit, Private Bag 65, Lilongwe, Malawi
Summary Operational research, within the context of a national disease control programme, may be described as the search for knowledge on interventions, tools or strategies which enhance programme effectiveness. There are two examples from Malawi of how operational research into recurrent tuberculosis and decentralization of treatment lead to information which enabled the National TB Control Programme (NTP) to change and improve its practice and policy. The key factors which allowed this to happen, and the guiding principles about integrating research into national programmes are discussed. TB programmes must have clear objectives, be able to identify constraints which prevent objectives being met and ask research questions around these constraints. There must be sufficient resources for research, both material and financial, and this requires programmes to incorporate a research agenda into their costed annual workplans. Training is also a key component of developing an integrated research programme, and should be included in the budgets. Research outputs should be judged in terms of activities undertaken and completed, papers written, and regular documentation of how research has influenced policy and practice. Finally, there should be strong advocacy for operational research, so that government policy makers can be convinced of its value. r 2003 Elsevier Science Ltd. All rights reserved.
Operational research studies in Malawi Recurrent tuberculosis Several clinical studies in sub-Saharan Africa have addressed the issue of recurrent tuberculosis (TB) after treatment has been completed. Several studies have found that recurrent TB is increased in patients infected with the human immunodeficiency virus (HIV).1–5 In the last 10–15 years Malawi, as with other countries in the region, has experienced a dramatic increase in HIV infection in the general population. There has been during this time period a corresponding increase in the proportion of TB patients found to be HIV-seropo*Corresponding author. C/o British High Commission, PO Box 30042, Lilongwe 3, Malawi. Fax: +265-772-657 E-mail address: [email protected] (A.D. Harries).
sitive, with rates in the last 2 years exceeding 75%.6 With rising HIV seroprevalence in TB patients, it might be expected that the number of patients with recurrent TB should increase. Malawi has collected good data on national notifications since 1984, and according to annual reports the number and percentage of patients with recurrent TB had remained fairly constant at between 3% and 5% between 1985 and 1997. This lead the Malawi National TB Control Programme (NTP) to suspect that patients with recurrent TB were being mis-registered under routine programme conditions. A country-wide operational research study was conducted in 1999, and patients who had been registered as ‘new cases’ in the TB register were interviewed about previous episodes of TB.7 A previous episode of TB was elicited in nearly 8% of 1254 patients who were being treated as new cases, particularly high rates of
1472-9792/03/$ - see front matter r 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S1472-9792(02)00061-6
144
mis-classification being found in patients with smear-negative pulmonary TB (PTB) (14%) and extrapulmonary TB (EPTB) (9%). The problem was discussed with all NTP staff at the national TB seminar in June 1999, and guidelines asking all patients carefully about previous episodes of TB and treating recurrent TB with a ‘step-up approach’ were distributed to all districts in the same year.8 In 2000, a similar country-wide research study was conducted, which found that rates of mis-classification had considerably reduced to less than 5% in patients with smear-negative PTB and less than 1% in patients with EPTB.8 In 2000 and 2001, the NTP recorded much higher numbers and rates of recurrent TB: 2057 recurrent cases (8% of all TB notifications) in 2000, and 2455 (9% of all TB notifications) in 2001.
Treatment of TB Between 1984 and 2001, all new patients registered with TB spent 1–2 months in hospital receiving the supervised initial phase of treatment.9 With TB case notifications rising from about 5000 in 1984 to nearly 28,000 in 2001, an adverse consequence of this strategy was marked congestion on TB wards where most patients received treatment. A country-wide survey in 1999 found a TB bed occupancy rate of 104%.10 The NTP decided in 1996 to try and tackle this problem and decentralize the initial phase of treatment, giving patients the option of remaining in hospital, attending health centres or receiving treatment at home under the supervision of family guardians. The treatment regimens were changed to full oral medication throughout.9,11 This new system was piloted under operational research conditions first in one district between 1996 and 1997,11,12 and then in a further 4 districts between 1998 and 1999. Data were collected on over 6000 patients, and showed good treatment success rates with about one-third of patients opting for guardian-based therapy and about one-quarter opting for health centre supervision (unpublished observations). Based on these findings, the NTP decided to expand the decentralized treatment country-wide. All districts in the Northern and Central Region of the country received training and implemented the new system in January 2001, and the Southern Region followed in July 2001. Preliminary findings of operational research currently being conducted in 2002 have found a TB ward bed occupancy rate of 58% (unpublished observations), confirming the perception that this approach has decongested the wards.
A.D. Harries
The key factors which enabled these studies to be completed and translated to policy and practice The key factors are shown in Table 1. The two studies both addressed constraints to TB control in the country. The NTP was puzzled that despite rising HIV-seroprevalence rates in TB patients, the number of recurrent cases of TB had remained stable. Therefore case notification of new and recurrent cases might be inaccurate. With regard to treatment, the NTP was concerned about the increasing congestion on TB wards, and wanted to decrease in-patient numbers. The studies were planned within the context of the NTP. There is an operational research programme embedded in the NTP with a dedicated budget line, and therefore readily available resources for carrying out research studies. There is one member of the central unit (not the TB programme manager) who is responsible for operational research, who sits on all the NTP management committees, who understands TB control at district level and who has competence in research methodology. There is a well functioning TB control programme with country-wide, standardized case finding, treatment and monitoring systems. Linked to this, there is a quarterly structured supervision programme conducted by experienced regional TB officers who try and ensure that records are accurate and well kept. This is crucial, as much of the research data is routine data collected within the programme in hospital and health centre TB registers, treatment cards, and patient identity cards. Finally, the NTP has developed the ability to move rapidly with its operational research programme. Research studies are usually undertaken according to workplans, and data are entered to computer software packages as studies progress. Once completed, data are quickly analysed and
Table 1 1. 2. 3. 4. 5. 6.
Key factors enabling operational research
Studies address constraints to TB control Studies are planned within the NTP There is a Central Unit officer responsible for operational research There is a well-functioning TB control programme The studies use established NTP systems There is an ability to move fast and make decisions on research findings which influence policy and practice
Operational research in tuberculosis control programmes
145
Box 1 NTP objective: to improve diagnostic practices *
Constraint
Using three sputum smears as a screening strategy for tuberculosis suspects is labour intensive, and there is a country-wide shortage of laboratory technicians
*
Question
Is there a simpler screening strategy? Is a strategy of screening pulmonary TB (PTB) suspects with two sputum smears as good as one using three sputum smears?
*
Research study
In one district in Malawi, compare a strategy of screening all PTB suspects using three sputum smears during 6 months with a strategy of using two sputum smears during 6 months 13
*
Findings
The smear-positive pick up rate in the laboratory and the pattern of TB, especially the proportion of patients with smear-positive PTB, was similar in the two study periods. The costs of the two sputum smear strategy was less than the cost of the three sputum smear strategy.
*
Implications
Expand and evaluate the two smear strategy in other districts with a view to country-wide implementation
translated into reports and papers, many of the papers being submitted to international peer reviewed journals. The NTP management group, consisting of the central unit and advisors drawn from projects/the medical school and districts, act on findings and are prepared to make decisions on control activities which are related to research findings. The credibility of research findings is enhanced if they have been written up as papers which have been accepted by international journals through the usual peer review process. This credibility is important when it comes to presenting the evidence-base and discussing policy changes with Ministry of Health and Population senior officials.
(i) is there a lack of knowledge about the issue in question? For example, what is the incidence of TB in prisoners? (ii) is there a lack of a suitable tool or can a better tool be used? For example, is secondary isoniazid preventive therapy for HIV-positive patients who complete anti-TB treatment useful in reducing recurrent cases of TB? (iii) are the tools used inefficiently or are the tools ineffective? For example, is it more cost effective to screen TB suspects using two rather than three sputum smears?13 (see Box 1)
Planning and resources
Guiding principles of integrating research into national programmes Objectives and constraints It is useful for an NTP to have clearly specified objectives and targets, and it can then look at the constraints at country level which hinder these objectives and targets from being met. Once constraints are identified, then research questions can be asked to better clarify the constraint or find a solution to the problem.
Research questions Research questions can be based around three main themes:
In Malawi, an annual costed workplan of activities is presented to the TB Programme Steering Group about 4–6 months prior to implementation. Part of this workplan includes the operational research programme for the forthcoming year. The workplan is usually approved by the Steering Group about 2 months prior to implementation, which enables the NTP to submit the research programme to the National Health Science Research and Ethics Committee before the year begins. There is need for a dedicated budget line for research, and part of this should include a generous miscellaneous flexible line item so that worthy research projects which may be developed and implemented during the year can be funded. There also needs to be some infrastructure in place if the NTP considers taking its research programme seriously (Table 2).
146
A.D. Harries
Table 2 * * * *
* *
* *
Office space Computer equipment, software, virus protection Transport to the field Stationary and communication systems, including email and possibly internet Training budget Research budget which covers research allowances for NTP staff Funds for reprints of articles Funds to allow local dissemination of research
Table 3 * *
* * *
Resources for research
Training in operational research
In service training District TB officers encouraged to develop own research Annual research training workshop Annual writing skills workshop Annual NTP review meeting to present research findings
Training In many resource-poor countries, NTP control staff will have little experience in operational research. Some degree of training is therefore important. This can be carried out in-country in a number of ways (Table 3). In-service training can be carried out whereby the Central Unit research officer implements simple studies in districts and closely monitors the collection of data with district TB officers. All NTP staff can be encouraged to develop their own studies, and the research officer can assist in drawing up protocols, analysing data and writing up results. In Malawi, an annual operational research training workshop is held for all NTP staff at which a generic protocol is developed. Research is then monitored in the field, and participants return 1 year later to discuss, analyse and write up research findings. Promising district staff and all central unit staff attend once a year a paper writing and critical appraisal skills workshop run by an editor of a Canadian Medical Journal. Finally, once a year the NTP holds a 2-day review meeting at which operational research findings are presented. All members of the NTP get the chance to present at these meetings, thereby learning the skills and developing the confidence of presentation to large audiences.
Judging the success of research Research is only useful if it delivers the goods. A research programme can be judged in relation to its annual workplan, and whether proposed outputs in terms of projects initiated, projects completed and papers written and published have been met. There must be an assessment about whether research findings have influenced policy and practice, while the ultimate proof of effectiveness is whether research has helped to improve programme performance. It is important to stress at this stage that if research is integrated within a programme, research activities must not interfere with the routine programme duties which should always take precedence if there is a clash of interests.
Advocacy Not everyone in the resource-strapped environment of developing countries will accept the importance of integrating research into a national programme. Strong advocacy from senior NTP staff and donor organizations may be necessary to convince the Ministry of Health and Population that placing resources for operational research is a worthwhile expenditure of funds.
Conclusion Operational research, within the context of a national disease control programme, may be described as research into strategies, interventions, tools or knowledge which enhance programme effectiveness. There is growing recognition that research in developing countries should be strongly linked to disease control,14 and that Africa in particular should have a strong say in setting relevant and appropriate research priorities.15,16 At present, much of the internationally published research being carried out in Africa is a result of collaborative ‘North– South’ partnerships, and the research agenda tends to be academic and reflects predominately Northern interests.17 Building a research agenda into a national programme is one way of countering this tendency. If programmes can develop a successful, initially small-scale, research programme this can be used as the stepping stone for establishing links with the local university, medical school and external research institutions. In this situation, programmes should become confident enough to develop their own research priorities, with the local university and external institutions joining in as equal partners rather than as dominant partners. The end products
Operational research in tuberculosis control programmes
of such research are more likely to be to the direct benefit of the local people.
References 1. Perriens JH, Colebunders RL, Karahunga C, et al. Increased mortality and tuberculosis treatment failure rate among human immunodeficiency virus (HIV) seropositive compared with HIV seronegative patients with pulmonary tuberculosis treated with ‘standard’ chemotherapy in Kinshasa, Zaire. Am Rev Respir Dis 1991;144:750–5. 2. Hawken M, Nunn P, Gathua S, et al. Increased recurrence of tuberculosis in HIV-1 infected patients in Kenya. Lancet 1993;342:332–7. 3. Elliott AM, Halwiindi B, Hayes, et al. , The impact of human immunodeficiency virus on response to treatment and recurrence rate in patients treated for tuberculosis: twoyear follow-up of a cohort in Lusaka, Zambia. J Trop Med Hyg. 1995;98:9–21. 4. Kelly PM, Cumming RG, Kaldor JM. HIV and tuberculosis in rural sub-Saharan Africa: a cohort study with two year follow up. Trans Roy Soc Trop Med Hyg 1999;93:725–7. 5. Sonnenberg P, Murray J, Glynn JR, et al. HIV-1 and recurrence, relapse, and reinfection of tuberculosis after cure: a cohort study in South African mineworkers. Lancet 2001;358:1687–93. 6. Kwanjana JH, Harries AD, Gausi F, Nyangulu DS, Salaniponi FML. HIV seroprevalence in patients with tuberculosis in Malawi. Malawi Med J 2001;12:7–10. 7. Harries AD, Hargreaves NJ, Kwanjana JH, Salaniponi FML. Relapse and recurrent tuberculosis in the context of a National Tuberculosis Control Programme. Trans Roy Soc Trop Med Hyg 2000;94:247–9.
147
8. Harries AD, Salaniponi FML. Recurrent tuberculosis in Malawi: improved diagnosis and management following operational research. Trans Roy Soc Trop Med Hyg 2001;95:503–4. 9. TB Manual of the Malawi National Tuberculosis Control Programme, 4th Ed. Ministry of Health and Population, Lilongwe, Malawi, 1999. 10. Harries AD, Kwanjana JH, Hargreaves NJ, Van Gorkom J, Salaniponi FML. Resources for controlling tuberculosis in Malawi. Bull WHO 2001;79:329–36. 11. Banerjee A, Harries AD, Mphasa N, et al. Evaluation of a unified treatment regimen for all new cased of tuberculosis using guardian-based supervision. Int J Tuberc Lung Dis 2000;4:333–9. 12. Manders AJE, Banerjee A, van den Borne HW, Harries AD, Kok GJ, Salaniponi FML. Can guardians supervise TB treatment as well as health workers? A study on adherence during the intensive phase. Int J Tuberc Lung Dis 2001;5:838–42. 13. Harries AD, Mphasa N, Mundy C, Banerjee A, Kwanjana JH, Salaniponi FML. Screening tuberculosis suspects using two sputum smears. Int J Tuberc Lung Dis 2000;4:36–40. 14. UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR). TDR Strategy 2000–2005. World Health Organization, 2000. TDR/GEN/ SP/00.1.Rev.1. 15. Costello A, Zumla A. Moving to research partnerships in developing countries. BMJ 2000;321:827–9. 16. Isaakidis P, Swingler GH, Pienaar E, Volmink J, Ioannidis PA. Relation between burden of disease and randomised evidence from sub-Saharan Africa: survey of research. BMJ 2002;324:702–5. 17. Wolffers I, Adjei S, van der Drift R. Health research in the tropics. Lancet 1998;351:1652–4.
Tuberculosis www.elsevierhealth.com/journals/tube
Integration of operational research into National Tuberculosis Control Programmes Anthony D. Harries* Malawi National Tuberculosis Control Programme, Community Health Science Unit, Private Bag 65, Lilongwe, Malawi
Summary Operational research, within the context of a national disease control programme, may be described as the search for knowledge on interventions, tools or strategies which enhance programme effectiveness. There are two examples from Malawi of how operational research into recurrent tuberculosis and decentralization of treatment lead to information which enabled the National TB Control Programme (NTP) to change and improve its practice and policy. The key factors which allowed this to happen, and the guiding principles about integrating research into national programmes are discussed. TB programmes must have clear objectives, be able to identify constraints which prevent objectives being met and ask research questions around these constraints. There must be sufficient resources for research, both material and financial, and this requires programmes to incorporate a research agenda into their costed annual workplans. Training is also a key component of developing an integrated research programme, and should be included in the budgets. Research outputs should be judged in terms of activities undertaken and completed, papers written, and regular documentation of how research has influenced policy and practice. Finally, there should be strong advocacy for operational research, so that government policy makers can be convinced of its value. r 2003 Elsevier Science Ltd. All rights reserved.
Operational research studies in Malawi Recurrent tuberculosis Several clinical studies in sub-Saharan Africa have addressed the issue of recurrent tuberculosis (TB) after treatment has been completed. Several studies have found that recurrent TB is increased in patients infected with the human immunodeficiency virus (HIV).1–5 In the last 10–15 years Malawi, as with other countries in the region, has experienced a dramatic increase in HIV infection in the general population. There has been during this time period a corresponding increase in the proportion of TB patients found to be HIV-seropo*Corresponding author. C/o British High Commission, PO Box 30042, Lilongwe 3, Malawi. Fax: +265-772-657 E-mail address: [email protected] (A.D. Harries).
sitive, with rates in the last 2 years exceeding 75%.6 With rising HIV seroprevalence in TB patients, it might be expected that the number of patients with recurrent TB should increase. Malawi has collected good data on national notifications since 1984, and according to annual reports the number and percentage of patients with recurrent TB had remained fairly constant at between 3% and 5% between 1985 and 1997. This lead the Malawi National TB Control Programme (NTP) to suspect that patients with recurrent TB were being mis-registered under routine programme conditions. A country-wide operational research study was conducted in 1999, and patients who had been registered as ‘new cases’ in the TB register were interviewed about previous episodes of TB.7 A previous episode of TB was elicited in nearly 8% of 1254 patients who were being treated as new cases, particularly high rates of
1472-9792/03/$ - see front matter r 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S1472-9792(02)00061-6
144
mis-classification being found in patients with smear-negative pulmonary TB (PTB) (14%) and extrapulmonary TB (EPTB) (9%). The problem was discussed with all NTP staff at the national TB seminar in June 1999, and guidelines asking all patients carefully about previous episodes of TB and treating recurrent TB with a ‘step-up approach’ were distributed to all districts in the same year.8 In 2000, a similar country-wide research study was conducted, which found that rates of mis-classification had considerably reduced to less than 5% in patients with smear-negative PTB and less than 1% in patients with EPTB.8 In 2000 and 2001, the NTP recorded much higher numbers and rates of recurrent TB: 2057 recurrent cases (8% of all TB notifications) in 2000, and 2455 (9% of all TB notifications) in 2001.
Treatment of TB Between 1984 and 2001, all new patients registered with TB spent 1–2 months in hospital receiving the supervised initial phase of treatment.9 With TB case notifications rising from about 5000 in 1984 to nearly 28,000 in 2001, an adverse consequence of this strategy was marked congestion on TB wards where most patients received treatment. A country-wide survey in 1999 found a TB bed occupancy rate of 104%.10 The NTP decided in 1996 to try and tackle this problem and decentralize the initial phase of treatment, giving patients the option of remaining in hospital, attending health centres or receiving treatment at home under the supervision of family guardians. The treatment regimens were changed to full oral medication throughout.9,11 This new system was piloted under operational research conditions first in one district between 1996 and 1997,11,12 and then in a further 4 districts between 1998 and 1999. Data were collected on over 6000 patients, and showed good treatment success rates with about one-third of patients opting for guardian-based therapy and about one-quarter opting for health centre supervision (unpublished observations). Based on these findings, the NTP decided to expand the decentralized treatment country-wide. All districts in the Northern and Central Region of the country received training and implemented the new system in January 2001, and the Southern Region followed in July 2001. Preliminary findings of operational research currently being conducted in 2002 have found a TB ward bed occupancy rate of 58% (unpublished observations), confirming the perception that this approach has decongested the wards.
A.D. Harries
The key factors which enabled these studies to be completed and translated to policy and practice The key factors are shown in Table 1. The two studies both addressed constraints to TB control in the country. The NTP was puzzled that despite rising HIV-seroprevalence rates in TB patients, the number of recurrent cases of TB had remained stable. Therefore case notification of new and recurrent cases might be inaccurate. With regard to treatment, the NTP was concerned about the increasing congestion on TB wards, and wanted to decrease in-patient numbers. The studies were planned within the context of the NTP. There is an operational research programme embedded in the NTP with a dedicated budget line, and therefore readily available resources for carrying out research studies. There is one member of the central unit (not the TB programme manager) who is responsible for operational research, who sits on all the NTP management committees, who understands TB control at district level and who has competence in research methodology. There is a well functioning TB control programme with country-wide, standardized case finding, treatment and monitoring systems. Linked to this, there is a quarterly structured supervision programme conducted by experienced regional TB officers who try and ensure that records are accurate and well kept. This is crucial, as much of the research data is routine data collected within the programme in hospital and health centre TB registers, treatment cards, and patient identity cards. Finally, the NTP has developed the ability to move rapidly with its operational research programme. Research studies are usually undertaken according to workplans, and data are entered to computer software packages as studies progress. Once completed, data are quickly analysed and
Table 1 1. 2. 3. 4. 5. 6.
Key factors enabling operational research
Studies address constraints to TB control Studies are planned within the NTP There is a Central Unit officer responsible for operational research There is a well-functioning TB control programme The studies use established NTP systems There is an ability to move fast and make decisions on research findings which influence policy and practice
Operational research in tuberculosis control programmes
145
Box 1 NTP objective: to improve diagnostic practices *
Constraint
Using three sputum smears as a screening strategy for tuberculosis suspects is labour intensive, and there is a country-wide shortage of laboratory technicians
*
Question
Is there a simpler screening strategy? Is a strategy of screening pulmonary TB (PTB) suspects with two sputum smears as good as one using three sputum smears?
*
Research study
In one district in Malawi, compare a strategy of screening all PTB suspects using three sputum smears during 6 months with a strategy of using two sputum smears during 6 months 13
*
Findings
The smear-positive pick up rate in the laboratory and the pattern of TB, especially the proportion of patients with smear-positive PTB, was similar in the two study periods. The costs of the two sputum smear strategy was less than the cost of the three sputum smear strategy.
*
Implications
Expand and evaluate the two smear strategy in other districts with a view to country-wide implementation
translated into reports and papers, many of the papers being submitted to international peer reviewed journals. The NTP management group, consisting of the central unit and advisors drawn from projects/the medical school and districts, act on findings and are prepared to make decisions on control activities which are related to research findings. The credibility of research findings is enhanced if they have been written up as papers which have been accepted by international journals through the usual peer review process. This credibility is important when it comes to presenting the evidence-base and discussing policy changes with Ministry of Health and Population senior officials.
(i) is there a lack of knowledge about the issue in question? For example, what is the incidence of TB in prisoners? (ii) is there a lack of a suitable tool or can a better tool be used? For example, is secondary isoniazid preventive therapy for HIV-positive patients who complete anti-TB treatment useful in reducing recurrent cases of TB? (iii) are the tools used inefficiently or are the tools ineffective? For example, is it more cost effective to screen TB suspects using two rather than three sputum smears?13 (see Box 1)
Planning and resources
Guiding principles of integrating research into national programmes Objectives and constraints It is useful for an NTP to have clearly specified objectives and targets, and it can then look at the constraints at country level which hinder these objectives and targets from being met. Once constraints are identified, then research questions can be asked to better clarify the constraint or find a solution to the problem.
Research questions Research questions can be based around three main themes:
In Malawi, an annual costed workplan of activities is presented to the TB Programme Steering Group about 4–6 months prior to implementation. Part of this workplan includes the operational research programme for the forthcoming year. The workplan is usually approved by the Steering Group about 2 months prior to implementation, which enables the NTP to submit the research programme to the National Health Science Research and Ethics Committee before the year begins. There is need for a dedicated budget line for research, and part of this should include a generous miscellaneous flexible line item so that worthy research projects which may be developed and implemented during the year can be funded. There also needs to be some infrastructure in place if the NTP considers taking its research programme seriously (Table 2).
146
A.D. Harries
Table 2 * * * *
* *
* *
Office space Computer equipment, software, virus protection Transport to the field Stationary and communication systems, including email and possibly internet Training budget Research budget which covers research allowances for NTP staff Funds for reprints of articles Funds to allow local dissemination of research
Table 3 * *
* * *
Resources for research
Training in operational research
In service training District TB officers encouraged to develop own research Annual research training workshop Annual writing skills workshop Annual NTP review meeting to present research findings
Training In many resource-poor countries, NTP control staff will have little experience in operational research. Some degree of training is therefore important. This can be carried out in-country in a number of ways (Table 3). In-service training can be carried out whereby the Central Unit research officer implements simple studies in districts and closely monitors the collection of data with district TB officers. All NTP staff can be encouraged to develop their own studies, and the research officer can assist in drawing up protocols, analysing data and writing up results. In Malawi, an annual operational research training workshop is held for all NTP staff at which a generic protocol is developed. Research is then monitored in the field, and participants return 1 year later to discuss, analyse and write up research findings. Promising district staff and all central unit staff attend once a year a paper writing and critical appraisal skills workshop run by an editor of a Canadian Medical Journal. Finally, once a year the NTP holds a 2-day review meeting at which operational research findings are presented. All members of the NTP get the chance to present at these meetings, thereby learning the skills and developing the confidence of presentation to large audiences.
Judging the success of research Research is only useful if it delivers the goods. A research programme can be judged in relation to its annual workplan, and whether proposed outputs in terms of projects initiated, projects completed and papers written and published have been met. There must be an assessment about whether research findings have influenced policy and practice, while the ultimate proof of effectiveness is whether research has helped to improve programme performance. It is important to stress at this stage that if research is integrated within a programme, research activities must not interfere with the routine programme duties which should always take precedence if there is a clash of interests.
Advocacy Not everyone in the resource-strapped environment of developing countries will accept the importance of integrating research into a national programme. Strong advocacy from senior NTP staff and donor organizations may be necessary to convince the Ministry of Health and Population that placing resources for operational research is a worthwhile expenditure of funds.
Conclusion Operational research, within the context of a national disease control programme, may be described as research into strategies, interventions, tools or knowledge which enhance programme effectiveness. There is growing recognition that research in developing countries should be strongly linked to disease control,14 and that Africa in particular should have a strong say in setting relevant and appropriate research priorities.15,16 At present, much of the internationally published research being carried out in Africa is a result of collaborative ‘North– South’ partnerships, and the research agenda tends to be academic and reflects predominately Northern interests.17 Building a research agenda into a national programme is one way of countering this tendency. If programmes can develop a successful, initially small-scale, research programme this can be used as the stepping stone for establishing links with the local university, medical school and external research institutions. In this situation, programmes should become confident enough to develop their own research priorities, with the local university and external institutions joining in as equal partners rather than as dominant partners. The end products
Operational research in tuberculosis control programmes
of such research are more likely to be to the direct benefit of the local people.
References 1. Perriens JH, Colebunders RL, Karahunga C, et al. Increased mortality and tuberculosis treatment failure rate among human immunodeficiency virus (HIV) seropositive compared with HIV seronegative patients with pulmonary tuberculosis treated with ‘standard’ chemotherapy in Kinshasa, Zaire. Am Rev Respir Dis 1991;144:750–5. 2. Hawken M, Nunn P, Gathua S, et al. Increased recurrence of tuberculosis in HIV-1 infected patients in Kenya. Lancet 1993;342:332–7. 3. Elliott AM, Halwiindi B, Hayes, et al. , The impact of human immunodeficiency virus on response to treatment and recurrence rate in patients treated for tuberculosis: twoyear follow-up of a cohort in Lusaka, Zambia. J Trop Med Hyg. 1995;98:9–21. 4. Kelly PM, Cumming RG, Kaldor JM. HIV and tuberculosis in rural sub-Saharan Africa: a cohort study with two year follow up. Trans Roy Soc Trop Med Hyg 1999;93:725–7. 5. Sonnenberg P, Murray J, Glynn JR, et al. HIV-1 and recurrence, relapse, and reinfection of tuberculosis after cure: a cohort study in South African mineworkers. Lancet 2001;358:1687–93. 6. Kwanjana JH, Harries AD, Gausi F, Nyangulu DS, Salaniponi FML. HIV seroprevalence in patients with tuberculosis in Malawi. Malawi Med J 2001;12:7–10. 7. Harries AD, Hargreaves NJ, Kwanjana JH, Salaniponi FML. Relapse and recurrent tuberculosis in the context of a National Tuberculosis Control Programme. Trans Roy Soc Trop Med Hyg 2000;94:247–9.
147
8. Harries AD, Salaniponi FML. Recurrent tuberculosis in Malawi: improved diagnosis and management following operational research. Trans Roy Soc Trop Med Hyg 2001;95:503–4. 9. TB Manual of the Malawi National Tuberculosis Control Programme, 4th Ed. Ministry of Health and Population, Lilongwe, Malawi, 1999. 10. Harries AD, Kwanjana JH, Hargreaves NJ, Van Gorkom J, Salaniponi FML. Resources for controlling tuberculosis in Malawi. Bull WHO 2001;79:329–36. 11. Banerjee A, Harries AD, Mphasa N, et al. Evaluation of a unified treatment regimen for all new cased of tuberculosis using guardian-based supervision. Int J Tuberc Lung Dis 2000;4:333–9. 12. Manders AJE, Banerjee A, van den Borne HW, Harries AD, Kok GJ, Salaniponi FML. Can guardians supervise TB treatment as well as health workers? A study on adherence during the intensive phase. Int J Tuberc Lung Dis 2001;5:838–42. 13. Harries AD, Mphasa N, Mundy C, Banerjee A, Kwanjana JH, Salaniponi FML. Screening tuberculosis suspects using two sputum smears. Int J Tuberc Lung Dis 2000;4:36–40. 14. UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR). TDR Strategy 2000–2005. World Health Organization, 2000. TDR/GEN/ SP/00.1.Rev.1. 15. Costello A, Zumla A. Moving to research partnerships in developing countries. BMJ 2000;321:827–9. 16. Isaakidis P, Swingler GH, Pienaar E, Volmink J, Ioannidis PA. Relation between burden of disease and randomised evidence from sub-Saharan Africa: survey of research. BMJ 2002;324:702–5. 17. Wolffers I, Adjei S, van der Drift R. Health research in the tropics. Lancet 1998;351:1652–4.